动态抓取中，双路径运动条件提升机器人响应能力
DynamicWAM: Dual-Path Motion Conditioning for World-Action Models in Dynamic Manipulation
- 用历史光流和运动参数双路径建模运动信息
- 在真实场景中达成46.7%平均成功率，超基线22.9个百分点
- 轻量化设计支持实时异步执行，适合复杂动态任务
动态操作要求机器人推断目标运动并快速响应，但现有世界-动作模型（WAMs）通常仅依赖当前帧，且同步运行大型骨干网络，限制了对运动的感知与实时控制。本文提出DynamicWAM，一种面向动态物体操作的紧凑型WAM，采用双路径运动条件机制。该模型引入历史光流条件，通过冻结预训练视频变分自编码器（VAE）编码时序对齐的光流帧，以保留空间运动结构；同时将位移、持续时间、速度和加速度等运动学描述符注入动作专家，提供运动幅度与时序信息。两条路径通过联合世界-动作注意力融合。结合轻量化压缩骨干与基于实时分块（RTC）的异步执行，实现快速响应。在DOMINO数据集上，DynamicWAM取得38.2%成功率和53.2分操作得分，优于所有对比基线。在12个涵盖线性、圆形及复合运动的真实任务中，平均成功率46.7%，较最强基线高出22.9个百分点。
原文摘要 · Abstract (English)
Dynamic manipulation requires robots to infer target motion and respond promptly, yet existing World-Action Models (WAMs) typically condition only on the current frame and execute large backbones synchronously, limiting motion awareness and responsive control in dynamic scenes. We propose DynamicWAM, a compact WAM for dynamic object manipulation with dual-path motion conditioning. DynamicWAM introduces history-flow conditioning, encoding temporally aligned optical-flow frames alongside the current observation through a frozen pretrained video VAE to preserve spatial motion structure, while injecting kinematic descriptors of displacement, duration, velocity, and acceleration into the action expert to provide motion magnitude and timing. The two complementary paths are fused through joint world-action attention. A distilled compact backbone and real-time chunking (RTC)-based asynchronous execution further enable responsive control. On DOMINO, DynamicWAM achieves a 38.2% success rate and a 53.2 manipulation score, outperforming all evaluated baselines. Across 12 real-world tasks spanning linear, circular, and compound target motion, it achieves a 46.7% average success rate, exceeding the strongest baseline by 22.9 percentage points.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。